← Back to Payment and Fintech

ACH Payment Processing

ACH payments through batch files, settlement windows, returns, and bank roles.

ACH, short for Automated Clearing House, is the main US network for low cost bank to bank transfers such as payroll, bill payments, account funding, and recurring debits. It is designed for high volume batch processing, not instant finality. That design keeps fees low, but it also explains why ACH behaves very differently from cards, wires, or real time payment rails.

Every ACH transfer begins with two banks and an authorisation. The party initiating the payment is the originator, such as an employer running payroll or a utility collecting a monthly bill. The originator works through an ODFI, the originating depository financial institution. The receiving bank is the RDFI. Before a debit is sent, the receiver normally must authorise it, because the ACH network has return rights for unauthorised or incorrect entries. Credits, such as direct deposit, are simpler operationally because the originator is pushing money out.

Most originators do not build raw ACH files by hand. They use payroll platforms, treasury software, or payment processors that collect routing numbers, account numbers, amounts, settlement dates, and transaction types, then produce NACHA formatted files. Those files are submitted to the ODFI before a cutoff time. Cutoffs matter because ACH moves in processing windows. Miss the window and the transaction usually shifts to the next batch.

The ODFI sends the batch to an ACH operator, either FedACH or the Electronic Payments Network. The operator sorts entries by receiving bank, calculates net obligations, and forwards the relevant entries to each RDFI. Settlement between banks happens through Federal Reserve accounts on scheduled timelines rather than on a per payment basis. That is one reason the network scales well for payroll and recurring payments.

When the RDFI receives its files, it posts credits or debits to customer accounts according to the effective entry date and its own posting schedule. For a payroll credit, that means employees see the deposit when the RDFI opens the file and updates balances. For a debit, such as a subscription pull, the customer may not notice until the bank posts the transaction and the available balance changes.

Settlement speed is often misunderstood. Standard ACH is not necessarily next minute and historically was often next day, but it can be same day if the originator submits within the operator windows and pays for Same Day ACH processing. Even then, same day does not mean irreversible. ACH has a return process. Entries can bounce for insufficient funds, closed accounts, invalid account details, or authorisation problems. Some return windows are short, while others, especially consumer unauthorised debits, can reach much further out.

That return model creates real operational constraints. Businesses using ACH debits need risk controls for account verification, fraud monitoring, and reserve management because a payment that looked successful in the morning can still be returned later. Payroll teams care about lead times because a late file can miss payday. Treasury teams care about prefunding and exposure limits set by the ODFI.

ACH is popular because it is cheap, widely connected, and reliable at scale. The tradeoff is that it asks businesses to think in batches, cutoffs, posting windows, and reversals. If you need immediate confirmation and near final settlement, ACH is the wrong tool. If you need to move huge volumes of ordinary bank payments with predictable operations, it remains one of the most important rails in the US financial system.